FYI a new release of our beloved HexaLab.net (the web-based hex-mesh vis tool) is online. We have just added the 27th dataset (totaling ~1k meshes) "Evocube: A Genetic Labelling for Polycube‐Maps" (doi.org/10.1111/cgf.14…). Thanks to Corentin Dumery for the pull-req!

In a paper to be presented at CVPR ( arxiv.org/abs/2311.10356 ), we propose a method that delivers realistic garment models from real-world images, even when the garments are loose instead of tight fitting. #deeplearning #computervision

Our paper written in collaboration with MBZUAI on modeling loose-fitting garments will be presented at SIGGRAPH. Crucially, we used guided diffusion models to synthesize the invisible parts while accurately modeling visible ones. arxiv.org/abs/2504.08353 #DeepLearning #Fashion

In a paper to appear at ICCV, we propose to use view-consistency distributions per-ray instead of fixed depth predictions to implement a sampling technique along the rays that implicitly regularizes the training of NeRFs. openreview.net/forum?id=ZQeKB…) #nerf

In a paper to appear at ICCV, we show how to count stacked objects such as those shown below, even though only those at the top are visible. Our key insight is that this suffices to estimate the number of objects per unit of volume. corentindumery.github.io/projects/stack…

The code and dataset for our ICCV25 Oral paper, “Counting Stacked Objects”, are now publicly available! 🔗We release datasets of 2D/3D stacked objects with ground-truth geometry and counts. Find out more on our project page: corentindumery.github.io/projects/stack…